Explore the vast range of titles available.

Abstract The impact of COVID-19 has been greatest in vulnerable US populations. This study examines the cumulative geographical and racial disparities of COVID-19 cases in nursing homes. Analysis of COVID-19 Nursing Home Data from Centers for Medicare & Medicaid Services was limited to weekly reports from the nursing homes that reported the ratio of black residents, from 2020-05-31 to 2021-01-17 (N=268,222 from 8,026 nursing homes). The outcomes were weekly COVID-19 cases and death per 1,000 occupied beds. Nursing homes were categorized by a geographic (rural vs. urban) and racial composition (>50% of residents are black vs. else). Elapsed time and county-level weekly COVID-19 cases and deaths/1,000 people were the key covariates. Multilevel zero-inflated negative binomial regression revealed evidence of cumulative COVID-19 disparity between rural and urban nursing homes. At the earliest time, COVID-19 incidence was lower in rural nursing homes than in urban nursing homes (IRR=0.406 for cases, 0.034 for death). The significant interaction with time implied that, over and above evolving disease prevalence, rural nursing homes became more likely than urban nursing homes to experience COVID-19 over time (IRR=1.057 for cases, 1.193 for death). Nursing homes, with >50% black residents, were more likely to experience COVID-19 than their counterparts at the earliest time (IRR=1.339 for cases, 5.630 for death), but independent of local disease prevalence, this disparity decreased over time (IRR=0.973 for cases, 0.972 for death). Our findings suggest that racial and geographic factors contribute to the cumulation of disadvantage during the COVID-19 crisis at the second half of 2020.

Abstract Across the lifespan, individuals adapt to change through the careful monitoring and adjustment of goals, demands, and performance-processes of self-regulation. Technology in support of self-regulatory processes may compensate for deficiencies in the ability to set, monitor, and work toward goals. Our purpose in this symposium is to forward the discourse on how health technology-from design to implementation-can assist older adults in their efforts to support their health and well-being in daily life. Our symposium begins with design considerations for technologies that support processes of information seeking, reflection, and action. Chin presents a process for designing conversation agents that guide dialogues with older adults to support informal self-regulated learning of health information. Nie and colleagues synthesize the literature on visual feedback to provide a framework that illustrates how visual design elements can link feedback to action. The symposium continues with papers that speak to older adults' experiences using technology to accomplish their goals. Mejía and colleagues use insight from older adults who had self-assessed their balance for 30 consecutive days to explore themes of self-monitoring accuracy and feedback preferences. Francis and colleagues use data from the Detroit-based Social Relations Study to illustrate how technology use and its implications vary when older adults engage with their weaker social ties. The symposium will conclude with a discussion led by Wahl, who will situate the papers, and the discourse on health technology design and application, within lifespan developmental and action perspectives on aging.

Abstract Relationship research has suggested that health among spouses is interdependent and should be considered jointly. Using data from the 2008/2010 and 2016/2018 waves of the Health and Retirement Study (3858 qualified couples; age=67.0±9.6), we investigated the joint influence of married partners' individual and shared cumulative biological risk on future health outcomes. Two risk indicators were constructed to indicate biological health in different domains. Individual grip strength, walk speed, lung function, and cystatin-C were biomarkers selected to construct frailty risk whereas blood pressure, pulse, waist circumference, C-reactive protein, glycohemoglobin, high-density lipoprotein cholesterol, and total cholesterol were biomarkers used to construct cardiometabolic risk. Shared risk was calculated as the number of risks the partners shared. We employed multilevel Poisson regression models to nest partners within couples and examine the effects of individual and shared cumulative risks on future functional limitations. Heckman correction was performed to correct potential selection bias. Our unadjusted models showed individual (frailty: b=0.22, p<.001; cardiometabolic: b=0.10, p<.001) and shared (frailty: b=0.17, p<.001; cardiometabolic: b=0.08, p<.01) risks are associated with greater future functional limitations. Further, shared cardiometabolic risk moderated the effect of individual risk (b=-0.01, p<.05). In the adjusted models, the direct associations between shared risks and future functional limitations were explained by indicators of partner selection and shared experiences. In the fully adjusted model, the cross-level interaction for frailty risk became statistically significant. The unique set of dynamics shown in our study offered new insights into understanding how couples influence one another in the context of multisystem biological health.

Abstract Falls are life-changing events in older adulthood. With an accurate understanding of balance, older adults can adapt to age-related changes in physical ability without prematurely restricting physical activity. The Daily Balance Project examines the implications of older adults' awareness of fall risk in daily life. For 30-consecutive days, following a fall-risk assessment, 40 older adults used a smartphone to report balance confidence and then perform four balance assessment and a 30-second sit-to-stand task to measure postural sway and fall-risk. Measures of postural sway showed greater intraindividual variability than balance confidence and fall risk. Multilevel models showed that awareness of balance fluctuated during the study and varied across individual differences in baseline fall-risk. Baseline fall risk also differentiated how balance confidence and postural sway were linked to subsequent momentary fall risk assessments. The findings are discussed within the framework of action-perspectives of adult development and awareness of aging.

Abstract Fall risk increases as older adults lose the functional resources necessary to maintain balance while completing everyday activities. As functional resources often decline gradually with age, momentary deficits may not be apparent until after a fall occurs. Mobile fall prevention technologies could support older adults in self-monitoring their ability to safely navigate their environments. In this paper we present perspectives on self-monitoring and feedback in a sample of older adults (n = 20, 50% female, age 65+) who had self-assessed their balance via a smartphone for 30 consecutive days. Thematic analysis of semi-structured interviews showed that fall history differentiated a) participants' awareness of day-to-day variation in functional ability; b) trust in the accuracy of self-monitoring; and c) imaginations of what types of feedback a mobile fall prevention technology should provide. Insight on older adults' internal self-monitoring processes and guidelines for feedback design are discussed.

Abstract The COVID-19 pandemic created an immediate, lasting impact on recruitment methods in academic research, most notably in the field of gerontology. To protect older adult participants' health during the COVID-19 crisis, the Daily Balance Project, a 30-day micro-longitudinal study of older adults' awareness of balance in daily life, shifted to complete remote administration. Our new remote protocol included developing new methodologies to recruit participants with varying degrees of fall risk and educational attainment. In this study, we present our approach to remote online recruitment and compare educational attainment, objective and subjective fall risk, and alignment of objective/subjective fall risk across three samples recruited via a) Fall Clinic registry (16 participants); b) University e-newsletter to faculty and staff (5 participants); c) social media recruitment (7 participants). Eligibility included being 65+ and wireless internet at home. For samples a and b, screening assessments were conducted via phone while baseline assessments were conducted in-person. For sample c, screener and baseline assessment were virtual. Analysis of recruitment methods aims to determine whether recruitment via social media platforms may provide a sample of participants with more variation in fall risk or alignment of subjective versus objective balance. Results demonstrate no significant differences in educational attainment (p=0.7949) or balance confidence (p=0.213), despite significant differences in the alignment of objective and subjective fall risk (p=0.031). Participants from samples a and b proved more able to accurately assess fall risk, while sample c had the most misalignment between subjective and objective fall risk assessments.

Abstract Balance confidence reflects one's estimate of their ability to maintain balance and avoid falls. Extensive literature has shown the relationships between balance confidence, functional limitations, and falls in later life. However, change in balance confidence, especially within short timescale, remains largely unknown and deserves further research. In this study, we aimed to investigate how older adults' balance confidence would change over 30 days and explore whether baseline characteristics would explain the individual differences in change. We used data from the Daily Balance Project that employed intensive-repeated measurements to examine the dynamics of subjective and objective fall risk across a month. Twenty-one participants (age=78.6±5.8, 48%female) were enrolled, and individual characteristics were measured upon recruitment. Throughout the study, participants self-reported their daily balance confidence using the Activity-Specific Balance Confidence (ABC) Scale. We performed growth modeling techniques to examine change within a multilevel framework. Our results showed that overall, ABC scores were high (79.9±17.4) at first, but the linear change was non-significant (b=0.03, SE=0.21, p=.89) on average. However, we found that balance confidence increased in individuals with higher educational attainment (b=0.37, SE=0.13, p<.01) and decreased among those with greater physical fall risk (b=-0.18, SE=0.07, p<.01) and accurate understanding of fall risk at baseline (b=-0.24, SE= 0.12, p=.04). Although ABC scores were stable within the period of one month, our study highlights the distinction of individual characteristics in the process of balance confidence appraisal. We suggest that these nuances should be taken into account when developing more fine-grained fall risk assessments and interventions.